The lifespan of an organism is the ultimate indicator of the molecular mechanisms active during aging of an organism. Short lifespan, well-established genetics, and conserved signaling pathways make the nematode Caenorhabditis elegans an attractive model for aging research. Large-scale automated screens for longevity genes in the nematode C. elegans often use bulk liquid culture combined with genetic or drug-induced blocking of progeny, introducing physiological stress on the animals. Common small-scale pilot screens adopt agar-based methods, which necessitate the tedious task of repetitively picking and transferring animals. In both approaches, it is challenging to add or remove reagents (e.g., food or drugs) at multiple time points during the lifespan, impairing detailed aging investigations. We report a compact and modular lifespan machine based on microfluidics that addresses the limitations of current methods. Our lifespan machine is comprised of three modules, (i) a microfluidic lifespan chip, (ii) a fluid exchange module and (iii) an imaging module. The microfluidic device houses crawling animals and an on-chip filter that allows the selective removal of progeny. The fluid exchange module controls the flow rate, duration and frequency of fluid injection for washing out progeny and for regularly feeding worms. A microcontroller is integrated into the system that is commanded by a simple GUI, allowing user-defined control of the fluid exchange module. The imaging module consists of a camera sensor and on-board Wi-Fi that captures, transmits and stores images at user-defined intervals. We designed custom software that analyzes the locomotory motion and conducts lifespan analysis. The entire system has a footprint of no more than 20 x 20
cm2 and has been designed such that it does not require any physical interaction, once the lifespan experiment is initiated. To reduce physiological stress on the animals, we have optimized the device geometry and feeding protocols such that C. elegans gait, body size, and lifespan are consistent with aging assays on agar. We find that the lifespan curves of wild-type and genetic mutants are consistent with the literature reports. In summary, the compactness and modularity of our approach allows building several of these systems, enabling highly parallelized cross-sectional and longitudinal aging experiments.